Tropical cyclone warning is one of the earliest weather warnings in Hong Kong and East Asia. The "typhoon warning" system was first introduced in Hong Kong in 1884, one year after the establishment of the Hong Kong Observatory. Over the years, there were several rounds of changes on the tropical cyclone warning system in Hong Kong. The latest change was in 1973 in which signals 5 to 8 were replaced by 8 NW, 8 SW, 8 NE and 8 SE respectively, leading to the current 1-3-8-9-10 scheme^[1]. Among these five signals, the Hurricane Signal No. 10 is certainly the most exciting and the rarest. Records since 1947 reveal that 12 typhoons necessitated the issuance of the Signal No. 10 in the past 65 years (1947-2011), on average about one in every 5 to 6 years.

Two most devastating storms in the past century in Hong Kong occurred in 1906 and 1937. Some 15 000 and 11 000 lives were lost respectively. Destructive storm surges were responsible for the heavy causalities in these two typhoon events. After the Second World War, Typhoon Wanda, which killed 183 people in 1962, claimed the first place in the hierarchy of deadliest typhoons in Hong Kong^[2]. To brave the storms, Hong Kong has over the years developed an effective warning system, and set up stringent building codes to protect the community. The implementation of various disaster prevention and mitigation measures has led to a steady decrease in the number of deaths/missing associated with tropical cyclones in the past 50 years.

Lives claimed (death or missing) by tropical cyclones in Hong Kong from 1960 to 2011.

Weather chart at 9:00 a.m. Hong Kong Summer Time on 1 September 1962 during Typhoon Wanda striking Hong Kong.

As mentioned in our previous blog article^[3], the tropical cyclone activity in the South China Sea was in a relatively quiet phase during the last two decades. The frequency of Signal No. 10 has decreased since the 1980s. The Hurricane Signal was issued only twice in the last three decades in 1983 (Ellen)^[4] and 1999 (York)^[5].

With slightly less frequent tropical cyclones affecting Hong Kong in the last few decades, some people may wrongly think that the typhoon risk in Hong Kong has declined. In fact, besides tropical cyclone visiting frequency, the risk of tropical cyclones in a place also depends on many other factors such as the infrastructure, topography, warning responses, social awareness and preparedness. A relatively quiet period in tropical cyclone activity may gradually lower society's awareness and preparedness on tropical cyclone threat, especially for the younger generation who has not experienced any major typhoon before. This may lead to a higher risk when a major storm eventually arrives. A recent study on tropical cyclone damage in Mainland China also showed that while Guangdong experiences far more landfalling tropical cyclones than other provinces like Zhejiang, Fujian and Guangxi, the average economic losses and casualties of a landfalling tropical cyclone in Guangdong is significantly lower than those in Zhejiang, Fujian and Guangxi. This is partly due to the fact that people in Guangdong are better prepared for typhoon strikes than those in other provinces where landfalling tropical cyclone is less common.

Although the Hurricane Signal No. 10 has not been issued for over a decade in Hong Kong now, it does not imply that the frequency of occurrence of such an event will not return to the long-term average of about one in every 5 to 6 years in the near future. Moreover, the recent disastrous tropical cyclone events in Taiwan, Japan and the Philippines triggered respectively by Typhoon Morakot (2009)^[6], Typhoon Talas (2011)^[7] and Tropical Storm Washi (2011)^[8] have clearly showed us the potential hazard of tropical cyclones to the coastal regions.

"Are we ready for the next Hurricane Signal No. 10?" This is the question we should ask ourselves prior to every typhoon season each year.



T C Lee & W H Lui


References:

[1] History of the Hong Kong Tropical Cyclone Warning Signals http://www.hko.gov.hk/informtc/tcsignal_history.htm

[2] Report of Typhoon Wanda 1962 http://www.hko.gov.hk/informtc/no10/wanda/wanda.htm

[3] Typhoon 5-6 on 3 April 2012 http://www.hko.gov.hk/blog/en/archives/00000120.htm

[4] Report of Typhoon Ellen 1983 http://www.hko.gov.hk/informtc/no10/ellen/ellen.htm

[5] Report of Typhoon York 1999 http://www.hko.gov.hk/informtc/90s/york/york.htm

[6] http://en.wikipedia.org/wiki/Typhoon_Morakot

[7] http://www.jma.go.jp/jma/en/typhoon_Talas.html

[8] http://en.wikipedia.org/wiki/Tropical_Storm_Washi_(2011)

This blog title may sound a bit peculiar to many people. "5-6" is certainly not a new name of tropical cyclone in the western North Pacific. In fact, 5 and 6 are the two most frequently occurred numbers of tropical cyclones affecting Hong Kong^[Note] each year, based on the statistics in the past five decades (See Figure 1).

Figure 1      Frequency distribution of annual number of tropical cyclones affecting Hong Kong from 1961 to 2010.

The tropical cyclone records in Hong Kong reveal that there were very large inter-annual and inter-decadal fluctuations in the annual number of tropical cyclones affecting Hong Kong. The lowest record is two in a year (in 1997 and 2007) while the highest record is 11 (in 1974). The average numbers of tropical cyclones for various periods are given in Table 1. While all the averages shown fall well within the range between 5 and 7, the 30-year average decreases from the 6 to 7 for the 30 year periods of 1961-1990 and 1971-2000 to between 5 and 6 for 1981-2010. The long term 50-year average from 1961 to 2010 is about 6.



Table 1      Average number of tropical cyclone crossing within 500km of Hong Kong and
average number of tropical cyclone signals for different periods.

The decrease in the 30-year averages in the last few decades is an interesting subject to look into. The inter-annual and inter-decadal variations in tropical cyclone frequency in the western North Pacific are likely related to a number of factors which affect the sea surface temperature and atmospheric circulations in the region. These factors include the El Niño and La Niña as well as the Pacific Decadal Oscillation^[1-2] which may have effects for periods of several years to decades. Moreover, some recent research suggested that the long term sea surface temperature rise due to human-induced climate change^[3] could also affect the tropical cyclone frequency, intensity and movement. Since the causes of the variations in the tropical cyclone frequency in the western North Pacific and the South China Sea are rather complex and may comprise both natural and human-induced influences, more studies are still required to further understand the relative contributions of natural variations (e.g. El Niño and La Niña) and human-induced factors to the past and future changes in tropical cyclone activity in the region.


Figure 2      Annual number of tropical cyclones (including tropical depressions) crossing
within 500 km of Hong Kong during 1961 to 2010.

In view of the presence of significant inter-decadal or even multi-decadal fluctuations in tropical cyclone activity in our region, the 30-year average of tropical cyclone counts may exhibit noticeable changes from one period to the other. Therefore, in depicting the norm for the annual number of tropical cyclone affecting Hong Kong, it may be more appropriate to refer to the long term average of 1961-2010 which is about 6 in a year.



T C Lee & Y Y Cheng


Note : Refer to tropical cyclones crossing within 500km of Hong Kong (22.3^o N, 114.17^o E). The use of 500 km range as a proxy indicator of tropical cyclones affecting Hong Kong is because the average number of tropical cyclones entering this range (about 6) is roughly the same as the long-term average of the annual number necessitating the issuance of local tropical cyclone warning signals in Hong Kong.


References:

[1] El Niño and La Niña http://www.hko.gov.hk/lrf/enso/enso-front.htm

[2] Pacific Decadal Oscillation http://www.hko.gov.hk/climate_change/pdo_e.htm

[3] Climate Change Webpage http://www.hko.gov.hk/climate_change/climate_change_e.htm

The International Telecommunication Union Radiocommunication Assembly 2012 (RA-12) deliberated on whether to change the definition of the universally adopted time standard "Coordinated Universal Time (UTC)" to the effect that it would decouple from the astronomical time scale and adhere strictly to the atomic time scale, thus dispensing with the need to apply leap seconds. The RA-12 decided to defer the decision and conduct further studies to ensure that all the technical options would be fully addressed. The outcome of these studies would be referred to the next Radiocommunication Assembly and World Radiocommunication Conference scheduled for 2015.

As the Earth's rotation is uneven, the astronomical time scale based on the Earth's rotation differs slightly from the atomic time scale based on the frequency of atomic oscillations: At present, one second on the astronomical time scale is slightly longer than that on the atomic time scale. Since 1972, to reconcile the two time scales, the UTC follows the atomic time scales ordinarily, but with a leap second introduced when necessary such that the difference between the UTC and the astronomical time scale is kept to less than 0.9 second (Figure 1).

In case the definition of the UTC is changed to follow the atomic time scale entirely, it will no longer be necessary to introduce leap seconds. The time scale UTC will then become continuous and predictable, thus simplifying the development, testing and maintenance of computer and navigation systems and reducing the chance of incidents. However, without leap seconds, the time of astronomical phenomena in a day would drift against the UTC at a very slow rate. It is estimated that the time of sunrise and sunset would become two or three minutes later than now in a hundred years, and about half an hour later in seven hundred years. Thousands or tens of thousands of years from now, it may occur that the sun would rise at 6:00 p.m., and set at 6:00 a.m.

Before any resolution of the issue, leap seconds will continue to be introduced to the UTC from time to time. For example, a leap second would be introduced immediately before 8:00 a.m. on July 1 (Hong Kong Time) this year.

Figure 1. Difference of UT1, an astronomical time scale, and UTC, Coordinated Universal Time. From 1972 till now, leap seconds have been added 25 times. Leap seconds were not introduced at regular time intervals. They were introduced 6 times, 8 times and 2 times in the 1980s, 1990s and 2000s respectively. It thus can be seen that the rate of rotation of the Earth would speed up and slow down from time to time. Furthermore, the definition of UTC was changed in 1972, hence the difference before and after 1972 could not be directly compared. (Source of Chart: International Earth Rotation and Reference Systems Service)